Cerebral Cavernous Malformations (CCM) are common vascular lesions of the central nervous system found in 1 in 200 people. Lesions consist of thin-walled enlarged vascular spaces prone to rupture leading to hemorrhagic strokes, seizures, and death. The disease can be hereditary, in which case affected individuals tend to have multiple lesions, placing them at even greater risk. A large population with hereditary CCM disease is found in New Mexico and the Southwestern United States in which the disease is caused by mutations in the gene, KRIT1. At present, treatment of the disease consists of choosing between neurosurgery to remove lesions, or conservative watchful waiting. Meaningful medical treatment of the disease will require an understanding of the underlying disease mechanisms, and will be greatly aided by animal models for use in pre-clinical trials of treatment. Recent studies have suggested a potential role for increased RhoA GTPase activity in the pathogenesis of the disease. The Ras family GTPases including RhoA, Rac1, and CDC42 regulate the cellular cytoskeleton, cell-cell interactions, and are involved in the cellular response to biomechanical stress. Other reports have suggested the importance of KRIT1 - Rap1 interactions in regulating cell junction proteins and endothelial cell apical - basal polarity. We hypothesize that KRIT1 is an important component of a mechanosensory apparatus that controls RhoA signaling with downstream effects on endothelial polarity. In this project we propose to use cell biology with physiologic levels of shear stress, and a newly developed animal model of CCM disease in mice to explore the role of KRIT1 in RhoA activation and cell polarity in vascular development and disease. We will describe the response of KRIT1 deficient endothelial cells when exposed to flow, as in living blood vessels, and study the response of signaling pathways in these cells. Using conditional gene targeting we will confirm that the underlying defect is found in the endothelial cells that line the CCM lesions. We will use an endothelial specific, drug inducible Cre system to knockout Krit1 from mice at birth. We have found that this approach leads to CCM lesions that can be seen and followed by MRI. We will use this system to test the efficacy of proposed treatments of CCM. In all of these aims we will evaluate the relative importance of RhoA activation and abnormal cell polarity. Further, we will develop and characterize non-invasive biomarkers of disease activity in CCM lesions, using the exaggerated progression of disease in this model and the access to tissue for mechanistic analysis to full advantage.